Loveland Cook, Cynthia A. PhD; Flick, Louise H. DrPH; Homan, Sharon M. PhD; Campbell, Claudia PhD; McSweeney, Maryellen PhD; Gallagher, Mary Elizabeth PhD
A basic premise of obstetric practice is to optimize pregnancy outcomes through preventive and ameliorative treatment. One area of clinical practice gaining increasing attention is the mental health of pregnant women and its effect on birth outcomes. Community prevalence studies estimate that 20–30% of all women experience at least one psychiatric disorder in a given year. Women of childbearing age have even higher reported rates.1–3 One psychiatric disorder affecting a disproportionate number of women of childbearing age is posttraumatic stress disorder, with lifetime rates ranging from 10.4% to 13.8%.4–7
People diagnosed with posttraumatic stress disorder usually have experienced or witnessed life-threatening traumatic events that elicit feelings of horror, terror, and fear.8 For women, the precipitating events most often are rape, childhood physical abuse, physical assault, or being threatened with a weapon.4,5,7 A large proportion of women experience trauma before the age of 25 years.6 Common symptoms of posttraumatic stress disorder include intrusive recollections of the traumatic stressor, avoidant/numbing behaviors, and hyper-arousal symptoms.8
Little research has focused on posttraumatic stress disorder in pregnancy to estimate either its prevalence or the likelihood of treatment for the disorder. Consequently, this study aims to estimate the prevalence of posttraumatic stress disorder in economically disadvantaged pregnant women, describe the proportion of women receiving treatment, and identify the associated risk factors that can facilitate screening for the disorder in clinical practice.
MATERIALS AND METHODS
Using a prospective cohort design, we recruited 744 pregnant Medicaid-eligible women at Women, Infants and Children Supplemental Nutrition Program sites in the city of St. Louis and in 5 rural counties in southeastern Missouri. Both areas have high levels of poverty and rates of infant mortality and low birth weight infants that exceeded national averages at that time. The sample was limited to black and white women, because they make up the vast majority of the population in both geographic locations. Trained research assistants administered a 2-hour in-person interview using the Diagnostic Interview Schedule9 and other study instruments. The Diagnostic Interview Schedule, a well known standardized diagnostic interview, assesses the presence of current and lifetime psychiatric diagnoses based on symptom, severity, and duration criteria in the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders-IV.8
Following approval of the study protocol by the Institutional Review Board, research assistants obtained informed consent from eligible pregnant women who sought services at their local Women, Infants and Children Supplemental Nutrition Program sites between February 2000 and August 2001. Our original plan was to sample participants from all the rural sites and representative urban sites. However, 2 sites in the city of St. Louis and 1 rural site refused or were unable to participate in the study. Replacement sites in the same geographical areas were selected based on their willingness to participate. All eligible women were then enrolled at each site in the order in which they were identified. State-level data on the number of pregnant women seen at the identified sites in the previous year overestimated eligible subjects for the study period, thereby precluding the random sampling of black and white subjects at each site. Eligible women were enrolled at each site until their numbers were proportional by race for women seen at the respective site.
Inclusion criteria for subjects, in addition to race, geographic location, and Women, Infants and Children Supplemental Nutrition Program enrollment at any point in their pregnancy, included being pregnant, having (or being eligible for) Medicaid coverage of health services, and being able to speak English. Mothers as young as 13 years old were included. Past interpretations of Missouri statutes allow pregnant minors to consent to medical care and participate in research without parental consent. Exclusion criteria included cognitive impairment that interfered with understanding of the interview questions. This was determined by having 12 or more errors on the dementia section of the Diagnostic Interview Schedule. Only one subject was ineligible because of cognitive impairment.
Of the total 878 women approached, 132 (15.0%) refused to participate. Ninety-three (11.0%) initially refused to enroll, 14 (1.6%) were unavailable for scheduled interviews, and 25 (2.8%) did not appear for their interviews. Of the 744 women who did participate in the study, 428 (57.5%) were black, and 316 (42.5%) were white (Table 1). Only 160 (21.5%) of the 744 women were married. More women resided in the rural Missouri (439 of 744, 59.0%) than in the city of St. Louis (305 of 744, 41.0%). Two of 5 mothers (41.8%) or 311 participants had not finished high school. Maternal age ranged from 13 to 43 years, with a mean of 22.3 ± 5.2 (standard deviation [SD]) years, and 171 of 744 (23.0%) enrollees were under 19 years of age. Despite their relatively young age, 192 of 744 (25.8%) reported having had at least one serious medical problem in their lifetime.
One of 5, or 161 of the 744 subjects, were interviewed during their first trimester of pregnancy. Another 303 (40.8%) were in their second trimester, and the remaining 279 (37.5%) were in their last trimester. Nine of 10 women (672 of 744) had their initial prenatal care visit in their first trimester, with another 56 (7.5%) first receiving care in their second trimester. Only 5 of 744 (0.7%) women received prenatal care for the first time during their last trimester, and 11 (1.5%) received no prenatal care at all. Nearly 60%, or 428 of 744 of the women, had experienced at least one previous live birth. Eighty-four (18.8%) delivered a previous pregnancy more than 3 weeks early.
Posttraumatic stress disorder, treatment for the disorder, and 18 other common psychiatric disorders were measured by using the fourth version of the Diagnostic Interview Schedule.9 This lay-administered standardized interview assesses diagnostic criteria in Diagnostic and Statistical Manual of Mental Disorders-IV, as well as symptom counts, onset and recency of symptoms, degree of disruption in work or social relationships, and treatment in the previous 12 months. The posttraumatic stress disorder module of the Diagnostic Interview Schedule begins with a list of 17 traumatic events, including being attacked or raped, experiencing combat conditions, seeing someone being seriously injured or killed, being threatened with a weapon, or being in a natural disaster. Subjects identify the worst event that ever happened to them and then respond to questions about posttraumatic stress disorder symptoms, age of exposure, onset of symptoms, remission, and treatment. In this study, posttraumatic stress disorder during pregnancy was based on symptoms occurring in the 12 months before and including the pregnancy interview. Treatment for posttraumatic stress disorder focused on whether or not subjects talked to a physician or other health professional about posttraumatic stress disorder–related behaviors or feelings in the past year.
Measurement of sociodemographic characteristics was based on items in the Diagnostic Interview Schedule. Items adapted from the Pregnancy Risk Assessment Monitoring System10 provided information on pregnancy history and environmental stressors. Developed by the Centers for Disease Control, this instrument assesses maternal health indicators related to prenatal care, attitude about pregnancy, pregnancy-related morbidity, living conditions, and stressors.
Data entry with verification and statistical analyses were conducted using SAS-PC 8 (SAS Institute, Cary, NC). We used a 5-step analytic strategy. First, we created descriptive statistics and summary profiles. Second, we calculated the prevalence of current posttraumatic stress disorder and assessed treatment for this disorder. Third, we evaluated risk factors for posttraumatic stress disorder using χ2 tests and Student t tests to identify any significant differences in sociodemographic, pregnancy, and environmental characteristics between women with and those without posttraumatic stress disorder. Fourth, we fitted logistic regression models to the data to determine the association between medical, environmental, and pregnancy-related factors and posttraumatic stress disorder. We tested our model for goodness of fit using the Hosmer and Lemeshow statistic. Finally, we calculated adjusted odds ratios (ORs) with 95% confidence intervals (CIs) to identify those risk factors that could be used as screening criteria to identify pregnant women with posttraumatic stress disorder in clinical practice.
Of the 744 women in this study, 101 (13.6%) had a diagnosis of posttraumatic stress disorder at some point in their lives. One in 13 women (57 of 744, 7.7%) had a current diagnosis of posttraumatic stress disorder. Another 0.9% (7 of 744) reported symptoms of posttraumatic stress disorder but did not meet the criteria of Diagnostic and Statistical Manual of Mental Disorders-IV for a current diagnosis. In comparison with other current psychiatric disorders examined, posttraumatic stress disorder was the third most common, following major depressive episode (80 of 744, 10.8%) and nicotine dependence (63 of 744, 8.6%).
Posttraumatic stress disorder is precipitated by exposure to one or more traumatic events. On average, the 57 women with current posttraumatic stress disorder had a mean of 4.9 ± 2.4 (SD) traumatic events over their lifetime. The most common events included the unexpected death of a close friend or relative (48 of 57, 84.2%), having something terrible happen to a close friend or relative (35 of 57, 61.4%), being sexually assaulted by a nonrelative (29 of 57, 50.9%), being mugged or robbed (26 of 57, 45.6%), experiencing a natural disaster (22 of 57, 38.6%), seeing someone killed or seriously injured (22 of 57, 38.6%), being sexually assaulted by a relative (20 of 57, 35.1%), and being in a serious accident (18 of 57, 31.6%). Twenty-one (36.8%) of the 57 women experienced the traumatic event that precipitated posttraumatic stress disorder before they were 15 years old.
The most commonly reported symptoms of posttraumatic stress disorder were intrusive distressing recollections of the trauma (57 of 57, 100.0%), psychological distress when exposed to cues resembling the trauma (55 of 57, 96.5%), difficulty concentrating (52 of 57, 91.2%), irritability or outbursts of anger (51 of 57, 89.5%), and avoidance of activities, places, or people associated with the trauma (51 of 57, 89.5%). They were somewhat less likely to report a sense of having a foreshortened future (31 of 57, 54.4%) or an inability to recall important aspects of the trauma (7 of 57, 12.3%).
Pregnant women with posttraumatic stress disorder reported moderate impairment in their daily functioning (mean 2.3 ± 0.8 [SD]), based on a scale ranging from 0 (none) to 4 (severe). Twenty-eight of the 57 women with current posttraumatic stress disorder (49.1%) reported difficulties with family, friends, and/or work during the same time period. Forty-one women (71.9%) reported 1 or more comorbid psychiatric disorders. Fourteen (24.6%) had 1 comorbid psychiatric diagnosis, 12 (21.1%) had 2, and another 16 (26.2%) had 3 or more. The most prevalent categories of comorbid diagnoses for women with posttraumatic stress disorder were mood disorders, followed by anxiety and substance-related disorders (Table 2). Of individual comorbid diagnoses, the most prevalent was major depressive episode (24 of 57, 42.2%). Other common diagnoses included manic episode (13 of 57, 22.8%), generalized anxiety disorder (11 of 57, 19.3%), nicotine dependence (10 of 57, 17.5%), and social phobia (10 of 57, 17.5%). Relatively few women reported alcohol abuse or dependence, despite its known association with posttraumatic stress disorder. On average, women with posttraumatic stress disorder in this study had 1.8 ± 1.7 (SD) comorbid diagnoses.
Only 7 of the 57 women with current posttraumatic stress disorder (12.3%) received treatment in the previous year for this disorder. One fourth of the women (15 of 57, 26.3%) wanted treatment for posttraumatic stress disorder but did not receive it. The remaining subjects (35 of 57, 61.4%) neither wanted nor received treatment for this disorder. Of the 50 women who did not receive posttraumatic stress disorder treatment in the previous year, 8 did receive treatment for another psychiatric diagnosis. There were no statistically significant differences in sociodemographic characteristics between women who did and did not receive treatment. However, women who received services had significantly more comorbid psychiatric disorders than those who did not receive services (mean 2.6 ± 2.0 [SD] and 1.5 ± 1.5 [SD], respectively; t = 2.26, P < .05). We found no significant differences between the 2 groups in level of impairment in the year before the interview.
The next analyses focused on identifying those characteristics associated with risk for posttraumatic stress disorder, including sociodemographic, environmental, and medical risk factors (Table 3). Although sociodemographic characteristics were not significantly different for women with and those without posttraumatic stress disorder, women with the disorder were significantly more likely to have had one or more serious medical illnesses in their lifetime and to have met the diagnostic criteria for major depressive episode, generalized anxiety disorder, drug dependence or abuse, and nicotine dependence. Pregnant women with posttraumatic stress disorder experienced significantly higher levels of life event stress and physical abuse in the previous 12 to 15 months than women without posttraumatic stress disorder. They also were significantly more likely to report separation from their mother as a child for more than 6 months and to have experienced multiple traumas in their lives.
A statistical model was developed to identify risk factors that would facilitate the clinical identification of pregnant women with posttraumatic stress disorder. By using logistic regression, risk factors were identified that significantly differentiated women with and without the disorder. As shown in Table 4, pregnant women with posttraumatic stress disorder had 5 times the adjusted odds of having a major depressive episode (OR 5.17; 95% CI 2.61, 10.26) and more than 3 times the adjusted odds of generalized anxiety disorder (OR 3.25; 95% CI 1.22, 8.62). Women with the disorder were more than 6 times as likely to have experienced 2 or more traumatic events in their lives (OR 6.61; 95% CI 1.97, 22.22). Subjects with posttraumatic stress disorder were nearly 2 times as likely to have been separated from their mothers for at least 6 months during their childhood (OR 1.89; 95% CI 1.01, 3.54). The Hosmer and Lemeshow goodness-of-fit test statistic was 1.06 with 7 degrees of freedom and a P value of .99. Because the statistic of 1.06 exceeds .05, we rejected the null hypotheses and concluded that the data fit the specified model.
In this study of economically disadvantaged pregnant women, posttraumatic stress disorder was the third most common psychiatric disorder, with a prevalence of 7.7%, closely paralleling the 8.1% reported earlier by Ayers et al.11 The lifetime prevalence of posttraumatic stress disorder (13.6%) also corresponds to that found in the general population of pregnant and nonpregnant women.4–7 Despite comparable rates in other studies, our findings may not be generalizable to pregnant women from higher socioeconomic levels or to women who are not black or white. Some research reports higher rates of posttraumatic stress disorder in low-income populations.12 Another factor that could influence generalizability is sampling from only urban and rural sites in a single state. Despite these limitations, posttraumatic stress disorder is common enough to be a clinical concern, particularly because the biological and psychological symptoms of this disorder may directly or indirectly affect birth outcomes. Breslau et al6 found that posttraumatic stress disorder significantly increased the probability of alcohol abuse and dependence. Although the use of alcohol may temporarily alleviate anxiety, promote sleep, and erase memories of trauma, its negative effect on fetal health is well documented.13 Neuroendocrine changes associated with chronic stress influence maternal–fetal health, including maternal vulnerability to hypertension and increased susceptibility to infection.14,15 Posttraumatic stress disorder may exert similar effects, although no known research has documented this relationship in pregnant women. However, research links high-risk behaviors to persons with posttraumatic stress disorder. Many of these behaviors, such as smoking, poor nutrition, and interpersonal violence, have known negative consequences for both pregnant women and their newborns.9,16 In a recent study, women with posttraumatic stress disorder had more complications of pregnancy, including more ectopic pregnancies, miscarriages, hyperemesis, and preterm contractions than their counterparts without posttraumatic stress disorder.17 The underlying mechanisms of how this disorder affects these outcomes are unknown.
Only 12.3% (7 of 57) of the women with posttraumatic stress disorder received treatment for this disorder. Seng et al16 suggest that women with abuse-related posttraumatic stress disorder may not seek mental health treatment but might be open to other forms of help. Among women who have been sexually abused, avoidance of reminders of the trauma may hinder their seeking needed health care services, including intrusive medical procedures in prenatal care. Yet it is the prenatal care setting itself that offers an ideal opportunity to identify pregnant women with posttraumatic stress disorder and make referrals for mental health treatment. Yehuda17 posits that traumatized persons with posttraumatic stress disorder are more likely to visit their primary care physicians than mental health professionals for treatment of symptoms.
Several factors are involved in the low treatment rates among pregnant women with posttraumatic stress disorder. Despite Medicaid coverage, mental health services are often limited in rural and inner city areas in this country. When services do exist, barriers to their access may be prohibitive, such as lack of transportation, inadequate child care, housing problems involving relocation, and long waiting times for appointments. Another factor may involve women's perceptions of their need for services. In this study, a large proportion of women reported they did not want treatment. Strong deterrents to mental health service use include hearing bad things about the care provided at a facility and fearing the stigma associated with mental health treatment.18 Another consideration is the often painful re-experiencing of trauma that can be inherent in the treatment of this disorder. Women with posttraumatic stress disorder also may have limited understanding of the value of mental health treatment, a disincentive cited by the New Freedom Commission on Mental Health.19 More likely, however, posttraumatic stress disorder may not be identified in prenatal care settings, and thus treatment referrals are not even made.
The importance of adequate screening and treatment of posttraumatic stress disorder during pregnancy is strongly supported in the literature. Prenatal assessments should detect those who need more extensive evaluation of posttraumatic stress disorder and provide treatment for the disorder. Greater awareness of symptoms related to this disorder, such as fear of pelvic exams, difficulty with reduction of tobacco or other substance use, and anxiety that seems disproportionate to presenting circumstances, will enhance more effective responses by health professionals. An informed approach to helping women with these problems is likely to increase compliance with prenatal care visit schedules and health-promoting behavior.
With the substantial overlap between symptoms of posttraumatic stress disorder and those of depression and anxiety disorders, health providers in both prenatal and primary care settings may miss the diagnosis of the disorder. In this study, some women with posttraumatic stress disorder received treatment, but it was for another psychiatric disorder. Optimal outcomes for women with posttraumatic stress disorder and co-occurring psychiatric diagnoses is associated with treating them simultaneously, rather than one after the other.20 For example, overlapping treatment for both posttraumatic stress disorder and the most common co-occurring diagnosis, depression, can include cognitive-behavioral therapy and antidepressant medications. Unique approaches to treatment for posttraumatic stress disorder, however, may also include eye movement desensitization and reprocessing and exposure therapy.
Identification of risk factors in this study demonstrated that women with posttraumatic stress disorder were 5 times more likely to have a major depressive episode, 3 times more likely to have generalized anxiety disorder, and more than 6 times more likely to have a history of multiple traumatic events. Screening for posttraumatic stress disorder based on multiple traumatic events is likely to contribute to the unnecessary reliving of these experiences. However, commonly used brief assessments for depression, spousal abuse, and domestic violence could be used to prescreen for the disorder. Research demonstrates that many obstetrician–gynecologists already conduct varying degrees of screening for depression in their practice.21 Women who are diagnosed with depression could then be evaluated for the presence or absence of posttraumatic stress disorder by using the screening tool developed and tested by Breslau et al.22 Comprising 7 questions on symptoms, the instrument identifies posttraumatic stress disorder with a sensitivity of 80% and specificity of 97% when using 4 or more symptoms as the cutoff score.
The high prevalence of posttraumatic stress disorder and low rates of treatment, whether from inadequate identification of the disorder in clinical practice, lack of knowledge about available treatment, or inaccessible mental health services, supports the provision of comprehensive treatment in prenatal care settings. Approaches to helping women with posttraumatic stress disorder include offering supportive counseling, teaching stress reduction techniques, initiating support groups, supporting continuity of care with the same provider, scheduling more frequent visits, and initiating nurse telephone calls between visits. Although women diagnosed with depression may also benefit from these treatment approaches, those with both posttraumatic stress disorder and depression may require additional mental health services. Ultimately, the benefit of detecting and treating posttraumatic stress disorder early in pregnancy is prevent or diminish its untoward physiological and psychological effects on mothers and their newborns.
1. Robins LN, Regier DA, editors. Psychiatric disorders in America: the epidemiologic catchment area study. New York (NY): The Free Press; 1991.
2. Kessler RC, McGonagle K, Zhao S, Nelson C, Hughes M, Eshleman, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Arch Gen Psychiatry 1994;51:8–19.
3. Narrow WE, Rae DS, Robins LN, Regier DA. Revised prevalence estimates of mental disorders in the United States: using a clinical significance criterion to reconcile 2 surveys’ estimates. Arch Gen Psychiatry 2002;59:115–23.
4. Kessler RC, Sonnega A, Bromet E, Nelson CB. Posttraumatic stress disorder in the national comorbidity survey. Arch Gen Psychiatry 1995;52:1048–60.
5. Breslau N, Davis GC, Andreski P, Peterson E. Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 1991;48:216–22.
6. Breslau N, Davis GC, Peterson EL, Schultz L. Psychiatric sequelae of posttraumatic stress disorder in women. Arch Gen Psychiatry 1997;54:81–7.
7. Resnick HS, Kilpatrick DG, Dansky BS, Saunders BE, Best CL. Prevalence of civilian trauma and posttraumatic stress disorder in a representative national sample of women. J Consult Clin Psychol 1993;61:984–91.
8. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.
9. Robins LN, Cottler LB, Bucholz KK, Compton WM, North CS, Rourke KM. Diagnostic interview schedule for the DSM-IV (DIS-IV). Available at: http://epi.wustl.edu/DIS/dishisto.htm
. Retrieved August 26, 2003.
10. Gilbert BC, Shulman HB, Fischer LA, Rogers MM. The Pregnancy Risk Assessment Monitoring System (PRAMS): methods and 1996 response rates from 11 states. Matern Child Health J 1999;3:199–209.
11. Ayers S, Pickering AD. Do women get posttraumatic stress disorder as a result of childbirth? A prospective study of incidence. Birth 2001;28:111–8.
12. Bassuk EL, Buckner JC, Perloff JN, Bassuk SS. Prevalence of mental health and substance use disorders among homeless and low-income housed mothers. Am J Psychiatry 1998;155:1561–4.
13. Day N, Leech SL, Richardson GA, Cornelius MD, Robles N, Larkby C. Prenatal alcohol exposure predicts continued deficits in offspring size at 14 years of age. Alcohol Clin Exp Res 2002;26:1584–91.
14. Culhane JF, Rauh V, McCollum KF, Hogan VK, Agnew K, Wadhwa PD. Maternal stress is associated with bacterial vaginosis in human pregnancy. Matern Child Health J 2001;5:127–34.
15. Wadhwa PD, Culhane JE, Rauh V, Barve SS. Stress and preterm birth: neuroendocrine, immune/inflammatory, and vascular mechanisms. Matern Child Health J 2001;5:119–25.
16. Seng JS. A conceptual framework for research on lifetime violence, posttraumatic stress, and childbearing. J Midwifery Womens Health 2002;47:337–46.
17. Yehuda R. Current concepts: post-traumatic stress disorder. N Engl J Med 2002;346:108–14.
18. Pignone MP, Gaynes BN, Rushton JL, Burchell CM, Orleans CT, Mulrow CD, et al. Screening for depression in adults: a summary of the evidence for the U.S. Prevention Services Task Force. Ann Intern Med 2002;136:765–76.
19. New Freedom Commission on Mental Health. Achieving the promise: Transforming mental health care in America. Final Report. DHHS Pub. No. SMA-03-3832. Rockville (MD): Department of Health and Human Services; 2003.
21. LaRocco-Cockburn A, Melville J, Bell M, Katon W. Depression screening attitudes and practices among obstetrician-gynecologists. Obstet Gynecol 2003;101:892–8.
22. Breslau N, Peterson EL, Kessler RC, Schultz LR. Short screening scale for DSM-IV posttraumatic stress disorder. Am J Psychiatry 1999;156:908–11.